From: Development of an automated speech recognition interface for personal emergency response systems
Scenario 1
Scenario 2
Scenario 3
Overall
Words read
36
108
Total Errors
4
6
0
10
Accuracy
89%
83%
100%
93%